Tamoxifen mimics the effects of endogenous ovarian hormones on repeated seizures induced by pentylenetetrazole in rats.

In the present study, the effects of tamoxifen on pentylenetetrazole (PTZ)-induced repeated seizures and hippocampal neuronal damage in ovariectomized rats were investigated. Thirty seven virgin female Wistar rats were divided to: (1) control, (2) sham-PTZ, (3) sham-PTZ-tamoxifen (sham-PTZ-T), (4) Ovariectomized -PTZ (OVX-PTZ) and (5) OVX-PTZ-tamoxifen (OVX-PTZ-T) groups. The animals of groups 3 and 5 were injected by tamoxifen (10 mg/kg) on 7 consecutive days. After 7 days of tamoxifen injection, they also were then injected by tamoxifen 30 min prior each PTZ injection. PTZ (40 mg/kg) was injected on 6 consecutive days and the animal behaviors were observed for 60 min. The histological methods were then used to determine dark neurons in hippocampus. A significant decrease in the seizure score was seen in OVX-PTZ group compared to Sham-PTZ. The animals of OVX-PTZ-T group had a significant higher seizure score compared to OVX-PTZ group. The dark neurons in DG of OVX group were lower than sham group (p<0.01). The numbers of dark neurons in CA1 area of OVX-PTZ-T group was higher than OVX-PTZ group (p<0.05) compared to control, the numbers of dark neurons in CA3 area showed a significant increase in Sham-PTZ and OVX-PTZ group (p<0.05 and p<0.01 respectively). Dark neurons in OVX-PTZ-T group were higher than OVX-PTZ group (p<0.05). It is concluded that pretreatment of the ovariectomized rats by tamoxifen increased PTZ-induced seizure score and dark neurons. It might be suggested that tamoxifen has agonistic effects for estrogen receptors to change the seizure severity.

INTRODUCTION

Estrogen has various effects on the nervous system and it seems to be an essential factor in the normal function of the brain [1, 2]. There is a wide range of estrogen receptors (ERα and ERβ) in numerous regions of the brain including hippocampus. Thus, it is suggested that this hormone affects the structures and functions of hippocampal neurons [3, 4]. This female sexual hormone acts via various mechanisms such as modulation of gene expression, regulation of neurotransmitter release, or direct interactions with neurotransmitter receptors [5]. Thus, estrogen can affect neuronal excitability and seizure susceptibility [6]. But there are controversial results in the case of the effects of estrogen on seizure. In previous studies, both the proconvulsant [7-9] and anticonvulsant effects [10, 11] of estrogen have been shown. We have previously shown that deletion of ovarian hormones attenuate seizure severity in a pentylenetetrazole (PTZ)-induced seizure model [9]. Furthermore, it has been shown that estrogen has neuroprotective effects against seizure-induced neuronal damage [10, 12, 13]. It is suggested that the different effects of estrogens on seizures may depend on various factors such as duration of the treatment, the priority of latency to seizure testing, the mode of administration, the applied dose of estrogen and hormonal status, the region of nervous system or the neurotransmitter system involved, the different seizure inducing models applied, and the sex of the samples [5].

Selective estrogen receptor modulators (SERMs) drugs such as ospemifene, tamoxifen and raloxifene depending on their chemical structure and the specific properties of the target tissues show both agonist/antagonist effects for estrogen actions [14, 15]. Tamoxifen is used for prevention and treatment of all stages of hormone-dependent breast cancer [16]. It is able to act as both estrogen agonist and antagonist depending on the used dose and tissue type. For example the maintenance of bone density and its cardio protective effects are examples of tamoxifen agonistic effects [17]. While, tamoxifen behaves as an antiestrogen in the breast tissue [18] and it also prevents of ovariectomy-induced downregulation of Bcl-2 and up-regulation of Bax expression [19]. In the central nervous system (CNS), tamoxifen was shown to reduce infarct size and neurobehavioral deficits due to middle cerebral artery occlusion in rats [20]. It also attenuates the microglial inflammatory responses and decreased the irradiation-induced brain damage [21]. Tamoxifen was also shown to have the estrogenic effects in hypothalamic differentiation during the neonatal period [22]. Moreover, it enhances choline acetyltransferase mRNA expression in a manner similar to the effects of estrogen in several basal forebrain regions [23]. Tamoxifen also inhibits the release of glutamate to prevent of the neuronal cell death [24]. It has also been suggested that tamoxifen acts as an antihormone in seizure and raises the electroconvulsive threshold in female mice [25].

Dark neurons were noticed to occur in neurosurgical biopsies and it was hypothesized that. They were produced by mechanical stress forces [26-28]. It has been reported that dark neurons may be produced under other conditions such as hypoglycemia, ischemia, and epilepsy [28-30] without any trauma or mechanical forces. Epilepsy produces widespread dark neurons throughout the brain, even under excellent fixation [30, 31].

Brain damage due to various kind of epileptic models such as temporal lobe epilepsy [32-34], partial [35] and febrile seizure [36] has been widely reported.

The aim of the present study was to investigate the effects of tamoxifen on pentylenetetrazole-induced seizure score and dark neuron production in ovariectomized rats.

MATERIALS AND METHODS

Animals and grouping

The experiments were carried out on 37 virgin female Wistar rats weighing 180-220 g. The animals were confined at random in metal cages, and maintained at the animal house under controlled conditions (12 hrs light and dark cycles, 21℃ and 50% relative humidity) with laboratory chow and water provided ad libitum. The animals were randomly divided into 5 groups (N=5-8 in each) as follows: 1- control, 2- sham-PTZ, 3- sham-PTZ-tamoxifen (sham-PTZ-T), 4- OVX-PTZ, 5- OVX-PTZ-Tamoxifen (OVX-PTZ-T).

Tamoxifen in a dose of 10 mg/kg [20] was used in this study. It was dissolved in saline plus a drop of tween 20 and administered on 7 consecutive days intraperitoneally (i.p). After 7 days of tamoxifen injection, PTZ (40 mg/kg) was injected intraperitoneally on 6 consecutive days [37] but the animals also were injected by tamoxifen 30 min prior to each PTZ injection. The rats of sham-PTZ and OVX-PTZ groups received normal saline plus tween 20 instead of tamoxifen. The animals of control group received normal saline (3 ml/kg, i.p.) instead of PTZ.

Surgeries

The rats were ovariectomized under ketamine (150 mg/kg, i.p) anesthesia [38]. After confirmation of anesthesia, ventral incision was made and ovaries and ovarian fats were removed. Ovaries were isolated by ligation of the most proximal portion of the oviduct before removal. These procedures were performed on the sham operated animals except, laparotomy was done without removing the ovaries. The animals were reversed to their cages to recover from the anesthesia [39]. After the surgery, the rats lived in animal house for 6 weeks before conducting the experiment for diminution of the endogenous sex hormones.

PTZ-induced repeated seizures

The animals were injected by 40 mg/kg PTZ [37]. After each PTZ injection, the rats were placed in Plexiglas cages separately and observed for 60 min to detect seizure score.

The resultant seizures were classified according to a modified Racine scale as follows: 1 - Mouth and facial movements; 2 - Head nodding; 3 - Forelimb clonus; 4 - Rearing; 5 - Rearing and falling.

Histological studies

Immediately after interventions and assessment of the animal behaviors, all rats were given a high dose of urethane and transcardially perfused with 100 ml of saline followed by 100 ml fixative solution (glutaraldehyde %3+paraformaldehyde %10 in 0.2 mol buffer phosphate at pH=7.4,T=40 C) for one hour [40, 41]. Just after perfusion, the brains were removed, processed according to routine histological methods and embedded in paraffin. Paraffin blocks were cut into serial coronal sections of 5 µm thickness. Next, ten uniforms random sample sections including hippocampus from each animal were chosen and mounted slides for toluidine blue staining.

Dark neurons per unit area (NA) of the CA1, CA3 and DG subdivisions of hippocampus were counted. The sections were examined under a light microscope using a ×40 objective lens (UPlanFI, Japan) and the images were transferred to a computer using a high-resolution camera (BX51, Japan). All sections were digitally photographed and the numbers of dark neurons were counted using a 10,000 µm2 counting frame. The mean numbers of neurons NA in different regions of hippocampus were calculated using the following formula [42, 43]:

Where "Σ"is the sum of counted particles appeared in sections, "a/f" is the area associated with each frame, "ΣP" is sum of frame associated points hitting space.

Statistical analysis

Data were expressed as mean±SEM and were analyzed by non-parametric statistical Kruskal-Wallis followed by Mann-Whitney tests. p values less than 0.05 were considered to be statistically significant.

RESULTS

The seizure scores in OVX-PTZ group were significantly lower than sham-PTZ group (p<0.05-p<0.01, Fig. 1). The animals of OVX-PTZ-T had a significant higher seizure score compared to OVX-PTZ group (p<0.05-p<0.001, Fig. 1) however, there was no significant difference between sham-PTZ-T group compared to sham - PTZ group in seizure score (Fig. 1).

Fig. 1

Comparison of seizure score of Sham-PTZ, Sham-PTZ-T, OVX-PTZ and OVX-PTZ-T groups. Data are presented as mean±SEM (n=8 in each group). The animals of both Sham-PTZ-T and OVX-PTZ-T groups were injected by tamoxifen (10 mg/kg) on 7 consecutive days. They also were then injected by tamoxifen 30 min prior each PTZ injection. PTZ (40 mg/kg) was injected on 6 consecutive days and the seizure score was recorded. The animals of both Sham-PTZ and OVX-PTZ groups received saline instead of tamoxifen. **p<0.01 and *p<0.05 compared to Sham-PTZ, †p<0.05, ††p<0.01 and †††p<0.001 compared to OVX-PTZ group.

The results of histological study showed that the mean of dark neuron numbers per unit area in DG of sham-PTZ group were higher than that of control group (p<0.001, Fig. 2). The produced dark neuron numbers in DG of OVX-PTZ group were lower than that of sham-PTZ group (p<0.01, Fig. 2). There was no significant difference in dark neurons of DG area between sham-PTZ -T and OVX-PTZ -T groups compared to sham-PTZ and OVX-PTZ groups respectively (Fig. 2). The mean of dark neuron numbers per unit area in CA1 of hippocampal subdivision in OVX-PTZ -T group was higher than that of OVX-PTZ group (p<0.05, Fig. 3). The produced dark neuron numbers per unit area of CA3 showed a significant increase in both Sham-PTZ and OVX-PTZ groups compared to control (p<0.05 and p<0.01 respectively, Fig. 4). The produced dark neurons in OVX-PTZ-T group were also higher than that of OVX-PTZ group (p<0.05, Fig. 4).

Fig. 2

(A) Comparison of dark neuron numbers per area in DG between control, Sham-PTZ, Sham-PTZ-T, OVX-PTZ and OVX-PTZ-T groups, the brain of control group was removed without PTZ injection. ***p<0.001 compared to control, ††p<0.01 compared to Sham-PTZ group. (B) Photomicrograph shows the dark neuron in DG area of rat hippocampus in the control, Sham-PTZ, Sham-PTZ-T, OVX-PTZ and OVX-PTZ-T groups. Toluidine blue staining, Arrow=dark neuron.

Fig. 3

(A) Comparison of dark neuron numbers per area in CA1areas among control, Sham-PTZ, Sham-PTZ-T, OVX-PTZ and OVX-PTZ-T groups, the brain of control group was removed without PTZ injection. &P<0.05 compared to OVX-PTZ group. (B) Photomicrograph shows the dark neuron in CA1 area of rat hippocampus in the control, Sham-PTZ, Sham-PTZ-T, OVX-PTZ and OVX-PTZ-T groups, Toluidine blue staining, Arrow=dark neuron.

Fig. 4

(A) Comparison of dark neuron numbers per area in CA3 area among control, Sham-PTZ, Sham-PTZ-T, OVX-PTZ and OVX-PTZ-T groups, the brain of control group was removed without PTZ injection. *p<0.05 and **p<0.01 compared to control, &p<0.05 compared to OVX-PTZ group. (B) Photomicrograph shows the dark neuron in CA3 area of rat hippocampus in the control, Sham-PTZ, Sham-PTZ-T, OVX-PTZ and OVX-PTZ-T groups, Toluidine blue staining, Arrow=dark neuron.

DISCUSSION

Animal and human studies have shown a clear relationship between seizures and pathological conditions in CNS [36, 44, 45]. Progressive functional and structural abnormalities in CNS, due to repeated seizures have been well documented [46-48]. It has been shown that seizure may lead to morphological changes such as production of dark neurons, in brain tissue.

Dark neurons, previously, were considered as histological artifacts in neurosurgical biopsies [27, 49] but later, the dark neurons were seen after brain trauma [26, 27]. Recently, it has been well documented that dark neurons are also produced without any trauma or mechanical forces [28, 50]. Dark neurons have basophilic appearance and morphological changes and might be seen after hypoglycemia, ischemia, stress and epilepsy [28-30, 51, 52]. Epilepsy has also been introduced as an important cause of dark neuron production [30, 31, 53]. The results of present study showed that PTZ induced repeated seizures were resulted in dark neuron production in hippocampal regions which confirmed our previous studies [43, 54]. Several studies have also confirmed hippocampal damages created by seizures [34-36].

In the present study, ovariectomy modified the PTZ-induced seizures. The results showed that the ovariectomized rats experienced seizures in lower scores than sham-operated ones. We have previously also showed that deletion of ovarian hormones lowered the score in repeated seizure model [37]. Latencies to the first minimal clonic seizure (MCS) as well the first generalized tonic-clonic seizures (GTCS) were also higher in OVX rats than sham operated ones when a high dose of PTZ was administered [9]. All of this findings support the proconvulsant effects of estrogen which has been repeatedly reported [7, 55, 56]. In contrast, it was shown that estrogen treatment may decrease the pilocarpine-induced temporal lobe epilepsy in animal model [57].

It has also been shown that estradiol and other female sex hormones have neuroprotective properties against seizure-induced neuronal damage [10, 12, 13]. However, the dark neuron numbers were higher in the sham group in comparison to the OVX group in the present study. It seems that the interactions between the sex hormones, seizure severity, and seizure-induced neurodegeneration is quite complex. In their study, Velísková et al. showed that estrogen postponed the kainic acid-induced clonic seizures. Estrogen also reduced kainic acid-neuronal damage in CA3 subfield and in the hilus of the dentate gyrus [10]. Hoffman et al. showed that 7 days after ovariectomy, administration of progesterone pellets (silastic capsules containing crystalline progesterone) had anticonvulsant effects, whereas estrogen pellets (0.1 or 0.5 mg/21 days) had little beneficial effect on seizure behavior. They also indicated that these two ovarian hormones reduced kainic acid-induced neuronal damage in the hippocampus [12]. Thus, in the present study, we would have predicted that loss of ovarian hormones should enhance PTZ-induced neuronal damage. In fact, the opposite effects were observed - for example loss of ovarian steroid hormones reduced hippocampal neuronal damage from PTZ-induced seizures. This observation might be due to higher intensity of seizure in the presence of ovarian hormones. Schauwecker et al. indicated that ovariectomy and chemically-induced ovarian failure resulted in a reduction of seizure-induced neurodegeneration; whereas, the administration of estrogen pellets immediately after ovariectomy, prevented the death of hippocampal neurons from kainate-induced seizures [58].

It is believed that the different effects of estrogens on seizures may depend on various factors such as the duration of the treatment, the priority of latency to seizure testing, the mode of administration, the applied dose of estrogen and hormonal status, the region of nervous system or the neurotransmitter system involved, the different seizure inducing models applied, and the sex of the samples [5].

Tamoxifen is a mixed estrogen agonist/antagonist, with estrogen antagonistic effects in some tissues such as breast [59] and estrogen agonistic effects in other tissues such as uterine [60] and bone tissues [61]. The results of present study showed that tamoxifen increased the PTZ-induced seizure score in ovariectomized rats however, it wasn't effective when was administered to the sham operated rats. It seems that the effect of tamoxifen is different in the presence and absence of ovarian hormones. Increasing in the uterine weight as well the luminal epithelial thickness has also been reported [62, 63]. This effect was weaker than 17 β-estradiol in ovariectomized rats. Based on the results of present study the estrogenic effects of tamoxifen in the brain might be suggested. Borowicz et al. who showed that tamoxifen at doses of 20-50 mg/kg significantly raised the threshold for electroconvulsions in female mice [25]. It has been also shown that tamoxifen impaired memory functions in experimental animals [64] however, in another study, Kathleen O'Neill et al. have shown that tamoxifen treatment had no effect on memory function and Alzheimer's disease [14]. The results of present study also showed that tamoxifen increased the dark neuron production in CA1 and CA3 areas of hippocampus. Interestingly, our results demonstrated that tamoxifen enhanced both the seizure severity and dark neuron density in ovariectomized rats. In fact, our results confirmed that tamoxifen can mimic the proconvulsant effects of ovarian hormones.

The results of present study showed that treatment of the ovariectomized rats with tamoxifen increased the seizure score and the dark neuron production due to repeated seizures. Regarding these results, the agonistic effects tamoxifen for estrogen receptors to change the seizure severity might be suggested.